© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 12679
    Prev
    Next
    INDEX
    Explanations

    words related to maintenance or upkeep

    oai_token-act-pair · gpt-3.5-turbo

    terms related to maintenance activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ologue
    -0.84
    inates
    -0.83
    ohydrate
    -0.82
    bern
    -0.82
    oche
    -0.75
    xxxxxxxx
    -0.75
    acters
    -0.75
    çīĪ
    -0.72
    izont
    -0.71
    onne
    -0.70
    POSITIVE LOGITS
    tenance
    1.10
     upkeep
    1.01
     maintenance
    1.00
     tasks
    0.84
     chores
    0.82
    taker
    0.82
     Maintenance
    0.79
     backlog
    0.78
     manuals
    0.77
     schedules
    0.77
    Activations Density 0.030%

    No Known Activations